Some Psuedorandom stuff
Let's say that you have 2 possible values for an 1bit PRNG.
The possible sequences: 0,1 and 1,0
What about 2bits? 16 from 00,01,10,11 to 11,10,01,00
What if the psuedorandom number generator ran for 256 unique 8bit charactors in a row, and then created a new group of 256 bytes every time it used them up?
How likely is this the next sequence to be accidentally the same? Would it be possible to break a 256byte substitution box with little work with a known plaintext? Known language/format? Unknown but redundant file?
Why is this 256byte key almost useless for text files even if totally random?
What I'm getting at is that all generators cycle but how bad is it and how easy is it to guess your current position (or even the key!) when there's single/multiple cycles.
Would it make a difference if you combined the first cycle with a second one (using the 2nd one to determine the rules for the first one in the current cycle)?
What if you just add values from the two cycles and they have 1 as the lowest common denominator?
Does the length of the total cycle double, add together, multiply, or what?
I'm sure this'll get the old nogging going.
Last fiddled with by nibble4bits on 20051231 at 15:46
